Estimating Yield from NDVI, Weather Data, and Soil Water Depletion for Sugar Beet and Potato in Northern Belgium

Crop-yield models based on vegetation indices such as the normalized difference index (NDVI) have been developed to monitor crop yield at higher spatial and temporal resolutions compared agricultural statistical data. We evaluated model performance of NDVI-based random forest for sugar beet potato farm yields in northern Belgium during 2016–2018. also whether weather variables root-zone soil water depletion growing season improved performance. The NDVI integral did not explain early late variability only partly explained sugar-beet variability. series crops were sensitive enough affecting conditions. found that water-saturated conditions elevated temperatures a large part late-potato combination with monthly precipitation, maximum temperature, farm-scale (R2 = 0.84, MSE 48.8) 0.56, 57.3) well from 2016 2018 Belgium.